0

I have an application insights instance and the "Availability" graph on the Overview page is showing zero availability. However, I have two app services, both of which are configured with health check URLs, and both of which are clearly being pinged and are returning 204s:

enter image description here

It feels to me like there's some kind of disconnection between an app service health check and app insights incorporating that health check into its availability stats. What might cause this?

me--
  • 1,978
  • 1
  • 22
  • 42

1 Answers1

1

By default, the availability tests check for status code 200, and only that. You can modify it by going to "availability" on your AppService.

https://learn.microsoft.com/en-us/azure/azure-monitor/app/monitor-web-app-availability

HTTP response

The returned status code that is counted as a success. 200 is the code that indicates that a normal web page has been returned.

edit

status code

Community
  • 1
  • 1
Alex AIT
  • 17,361
  • 3
  • 36
  • 73
  • Thanks for your answer. OK, wow. I wasn't even aware I had to set this up in AI. The way I did it was by adding a `health_check_path` in my app service, not in AI (I'm using Terraform). Terraform's docs link to [this](https://github.com/projectkudu/kudu/wiki/Health-Check-(Preview)), which now feels like a different thing to me than AI health checks. Very confused. – me-- Jun 18 '20 at 03:03